PC connected to UPS, hangs w/o display on battery mode

[Sempron Guy]

If the UPS is inadequate it usually turns off right? In my case, I tried two UPS units already, the first one is an Intex 1000va. My pc listed on my profile is just around 2 mos. old and in the first month during back up mode my UPS works fine, it stays powered on and I can shut it down properly. There are even instances where my pc is doing BOINC and the UPS handles it fine as expected since total load is probably just around 350w or less. Now a few weeks ago, power interruptions starts to become rampant again and it was then that the issue started to appear. During backup mode and usually under idle loads so probably less than 200w incl. the monitor, it works fine. But once power resumes, the display turns off, but the computer stays on, and I can see that it's still running with the lights and fans still spinning but no response on the keyboard (pressing the capslock button, num lock). I made sure it's not the monitor acting up cause I tried it without the monitor and keyboard still locked up. Monitor also turns on fine connected alone with the UPS. So at that point I thought maybe it's the UPS. Pulled an APC BackUPS rated at 625va/300w from our family pc hooked it up on my computer at idle pulled the UPS plug from power source and same issue. It backs up the pc fine but once power resumes, Lights and fans keep on spinning, display turns off, keyboard unresponsive. I need help cause I am thinking off buying another UPS and don't want to waste cash if it turns out with the same issue.

 

[newtekie1]

I believe both of those UPS units use the cheaper square wave method of output. It could be that your PSU just doesn't like the square wave output, I've seen that before. Get a good UPS unit with a simulated sinewave.

 

[Sempron Guy]

ok sorry for the mix up but I remembered wrong. Tried it right now and the UPS backs up the computer fine, but once power resumes back the display turns off along with the symptoms I mentioned above. I would also like to note that does that on random as I tried it now for the 2nd time and everything works fine.

 

[Bill_Bright]

Any decent PSU should easily be able to run just fine with simulated or "approximated" sinewave outputs. I mean they did so for decades and PSUs have improved greatly since then. Pure sinewave outputs are only a recent phenomenon that became more popular simply because the prices came down. The only reason approximated sinewave outputs get bad press is ONLY due to aggressive and misleading marketing by those selling pure sinewave UPS.

Not sure I have ever seen a UPS with a real square wave output. Even budget ones shape the output to a simulated or "approximated" sinewave.

I know nothing about Intex UPS but APC is my preferred brand. But note there are budget models and like power supplies, you should use the better UPS on your computers. So, I always look for an UPS with AVR - automatic voltage regulation.

It should also be noted that the ATX Form Factor standard only requires ATX power supplies to "hold up" power for a mere 16ms should input power drop below a set threshold (I think it is 90VAC). 16ms is much faster than you can see (in lights flickering, for example). But sadly, not all power supplies meet that requirements so a good UPS with AVR must be able to react much quicker - no more than 10ms is good.

And lastly UPS batteries only last 3 - 5 years and need to be replaced. If weak, they will not allow for a clean, if any cut-over.

 

[Sempron Guy]

the Intex UPS i got has a built in avr on it. The APC doesn't. They are both less than a year old. In fact the intex was bought last march. I'm on a loop here. Both UPS works fine but the computer goes nuts when power resumes. Btw I connected a secondary monitor connected directly to power source not on the ups. Unplugged the ups it backed up the cmputer fine. Once I plugged it back on the same no display on the secondary monitor so yeah PC just hangs.

 

[OneMoar]

disconnect the usb/serial from the ups to the pc and cycle the wallpower and see if it still occurs
might be a issue with the auto shutdown procedure or a bad communication driver

 

[Sempron Guy]

disconnect the usb/serial from the ups to the pc and cycle the wallpower and see if it still occurs
might be a issue with the auto shutdown procedure or a bad communication driver

these UPS I got are pretty basic doesn't have those connectors.

 

[OneMoar]

a UPS without a usb connection ?
then its not a UPS
there should be a mini USB B on the back for communication from the ups to the pc

 

[Sempron Guy]

it's a basic UPS that's just it

 

[OneMoar]

no communication means its not a UPS its a battery backup and it also tells me that the unit is overall garbage

I have 10 year old APC's that have usb serial AND ethernet

 

[Bill_Bright]

a UPS without a usb connection ?
then its not a UPS

It does not have to have a USB connector to be an uninterruptible power supply. If it does not come with a USB connection for monitoring, it is just is not designed to be connected to or monitored by a computer.

 

[Sempron Guy]

no communication means its not a UPS its a battery backup and it also tells me that the unit is overall garbage

I have 10 year old APC's that have usb serial AND ethernet

blame it on marketing i suppose or perhaps inform APC about it? Cause it's listed under their UPS secton on their site and it says on the box it's a ups. Anyway we are off the main topic.

 

[Bill_Bright]

It is tunnel vision to assume UPSs are just for computers. They have been around much longer than the PC and used for many things other than computers, including burglar alarm systems, garage door openers, home theater systems, health monitoring and life support systems, security lighting, refrigerated medicine storage - all sorts of things that don't need monitoring by a computer.

The lack of a USB port for computer monitoring IN NO WAY suggests an inferior product.

 

[OneMoar]

defective battery backup unit is defective
./thread

 

[Sempron Guy]

the APC UPS works fine on the family pc and isn't showing any of the issues I mentioned. Though the specs overall consumes lesser power. I don't it matters though as the issue is not backing up but once the power restores to normal.

 

[OneMoar]

its probly the change in wave-form going from stepped sine to full sine some psu's just don't like that
replace it with a good ups before you fry something

 

[Sempron Guy]

weird cause it worked fine the first month or so what changed?

 

[newtekie1]

Any decent PSU should easily be able to run just fine with simulated or "approximated" sinewave outputs. I mean they did so for decades and PSUs have improved greatly since then. Pure sinewave outputs are only a recent phenomenon that became more popular simply because the prices came down. The only reason approximated sinewave outputs get bad press is ONLY due to aggressive and misleading marketing by those selling pure sinewave UPS.

Not sure I have ever seen a UPS with a real square wave output. Even budget ones shape the output to a simulated or "approximated" sinewave.

I know nothing about Intex UPS but APC is my preferred brand. But note there are budget models and like power supplies, you should use the better UPS on your computers. So, I always look for an UPS with AVR - automatic voltage regulation.

It should also be noted that the ATX Form Factor standard only requires ATX power supplies to "hold up" power for a mere 16ms should input power drop below a set threshold (I think it is 90VAC). 16ms is much faster than you can see (in lights flickering, for example). But sadly, not all power supplies meet that requirements so a good UPS with AVR must be able to react much quicker - no more than 10ms is good.

And lastly UPS batteries only last 3 - 5 years and need to be replaced. If weak, they will not allow for a clean, if any cut-over.

While I do mostly agree with you. I have seem some power supplies that just don't like squarewave output. Usually this, in my experience, is with unit that have been switched to and from battery backup a lot. This stresses out the PSU and eventually they just wear out.

It could also be the hold-up time of the PSU going down as the main cap wears out too. This is another issue with cheap UPSes. Their switchover time is usually higher, putting more stress on the components of the PSU.

 

[Bill_Bright]

I have seem some power supplies that just don't like squarewave output.

I have too, just not with upper tier PSUs. And while frequent cut-overs could indicate very unstable mains, it could also indicate a poor facility ground to Earth ground and/or the sensitivity of the UPS set too tight - those are not PSU faults.

This is another issue with cheap UPSes.

Did you mean to say UPS?

I note even that relatively low-end APC Back-UPS 650, which can be had for under $75 if you shop around, has a "typical" transfer time of just 8ms and a rated maximum time of 12ms. That is plenty fast for any PSU that complies with the ATX standard hold-up requirement of >16ms.

 

[newtekie1]

Did you mean to say UPS?

Yes. The transfer time might be under the ATX spec, but they are longer than the more expensive units. Causing excess wear on the PSU over multiple switch overs. It isn't really an issue in an area that might switch to battery once or twice a year. However, it can be an issue in areas that switch over once or twice a month or more. The strain of keeping the computer running with no incoming power can wear out the components of the PSU.

This is a problem with the cheaper UPSes, not a PSU issue.

 

[Bill_Bright]

While I agree about the wear and tear, I believe the wear and tear from such unstable mains is harder on electronics plugged directly into the - especially since such unstable events are typically accompanied with excessive surges and spikes, dips and sags - which are even harder on power supply regulator circuits. Otherwise, why use a UPS in the first place?

 

[FordGT90Concept]

Have you gone into the BIOS and changed the power loss setting to "power off?" Default is usually "resume previous." You might be experiencing a conflict between UPS, operating system, and BIOS settings.

I do...
BIOS: power off
OS: disable sleep states, power off monitor with idle time
UPS software: 2 minutes on AC, power off.

UPS tells OS to shutdown after 2 minutes then turn itself off -> BIOS says to stay off. When power is reliably back on, I turn UPS on, then computer on. Computer is none the wiser that there was no power in that period.


I think what might be happening is that your PSU/motherboard thinks it is in a standby power state when power restores and it's not providing enough 12v for the CPU to come out of that power state so it's stuck there until it loses all power resetting the current state.

 

[OneMoar]

Have you gone into the BIOS and changed the power loss setting to "power off?" Default is usually "resume previous." You might be experiencing a conflict between UPS, operating system, and BIOS settings.

I do...
BIOS: power off
OS: disable sleep states, power off monitor with idle time
UPS software: 2 minutes on AC, power off.

UPS tells OS to shutdown after 2 minutes then turn itself off -> BIOS says to stay off. When power is reliably back on, I turn UPS on, then computer on. Computer is none the wiser that there was no power in that period.

is his 'ups' has no software no pc communication
crappy unit is my guess

 

[FordGT90Concept]

If it is an Intex 1000VA Gaama, Amazon says it has "MANAGEMENT Optional USB/RS-232 Port Supports Windows ® 98/2000/2003/XP/Vista/2008" which is typical for UPSs.

I think the BIOS power setting on AC loss is the key to the problem more so than anything else. Updating motherboard BIOS could help resolve that issue too if they knew about it and fixed it.

 

[OneMoar]

IF it is an Intex 1000VA Gaama, Amazon says it has "MANAGEMENT Optional USB/RS-232 Port Supports Windows ® 98/2000/2003/XP/Vista/2008" which is typical for UPSs.

I think the BIOS power setting on AC loss is the key to the problem more so than anything else. Updating motherboard BIOS could help resolve that issue too if they knew about it and fixed it.

the machine isn't loosing power tho if there is enough of a voltage drop to trip the PSU's UV protection then the machine would power OFF its not its just hanging up fans running ect ect